Information system of actors, operation types and operations, generates an optimal schedule of operations and evaluates its metrics.

Functions

• generate an optimal schedule of operations
• evaluate metrics of the schedule
• extract schedule
• extract scheduled operations for a specific actor
• create, update, delete a list of operations
• delete unscheduled operations
• create, get all, get by id, update, delete by id actor, operation type, operation

Optimisation

It is assumed there is a constraint that only one actor may conduct a given operation on a certain day. Therefore there is a problem to find a uniform distribution of operations among actors. The optimisation process consists of finding all possible schedules and selecting one with minimum mean deviation of operation count per actor from its mean value.

Optimisation Service

Database

SQLite database schema
Taking into consideration a very small number of database objects it was decided to use JDBC and connection pool to increase performance instead of Spring Data or any other ORM framework. The system is intended to have only one user who modifies the data and not a big number of readers, thus SQLite was selected as database engine.

Technologies and tools

• Java 8, IntelliJ IDEA
• SQLite, SQLite Studio
• JDBC, Apache DBCP (connection pooling framework)
• Spring framework (Boot, Web/MVC, Test, Actuator, SLF4J/Logback, caching, Swagger 2 API documentation)
• Maven, Git
• VisualVM (Java profiler), Apache JMeter, Insomnia REST Client

Future improvements

• accelerate schedule generation with multithreading
• devise an approximate, heuristic solution, probably using dynamic programming
• design UI, probably using React framework
• provide performance benchmarks
• conduct profound static code analysis
• introduce security layer
• add integration tests